An easy way to perform a radiation damage calculation in a complicated geometry

With the advent of ITER the need for more and more accurate fusion neutronics analyses increases. The standard software package for these analyses is MCNP, by which 3- dimensional radiation transport analyses may be performed. Demands on requested parameters are expanding from the more traditional neutron flux values to all kinds of derived parameters. These include radiation heating and reaction rates. Also for components receiving a high neutron flux the calculation of radiation damage is an important issue. Usually approximations are made to calculate the radiation damage. In this paper it is shown that with negligible additional effort an almost exact analysis is possible in arbitrary complicated situations. A correct estimation of the expected radiation damage of components involves the use of molecular dynamics simulations. However, this is only possible for scientific investigations due to the extremely long calculation times needed for actual engineering studies. Hence, one nearly always resorts to using pre-compiled damage cross-section data, which are used as response cross sections in the MCNP analyses. One often needs an evaluation of the radiation damage in a material (e.g. stainless steel). Therefore in the current approach one frequently uses pre-calculated damage cross-section data for a few specific materials in a relatively coarse energy group structure (e.g. the 640-group SAND-II structure such as used in the DAMSIG library). In this approach the details of the analysis are only taken into account to a limited extent. Important factors can only be approximated. These include the composition of materials (is in the actual problem the material composition identical to the one on the library?) the effect of temperature (is the actual temperature identical to the one on the library?) and selfshielding (the use of group cross-section data always leads to approximations). In the current approach continuous-energy damage cross-section data are used, which were generated by the nuclear-data processing code NJOY. The data are available to MCNP as response cross sections. A damage calculation in MCNP for a material mixture involves many isotopes, which makes the standard method in MCNP completely impracticable due to the large amount of of pre- and post-processing. In this paper it is shown, that these continuous-energy damage cross-section data may be used in a much simpler way by using a modified material specification, which is weighted by the damage cross section of the isotopes. A theoretical foundation for this approach is given. The data are compared with damage cross-section data from literature, showing a good agreement. It is demonstrated that, without additional effort, a calculation of radiation damage is possible which is completely consistent with the underlying radiation transport calculation. This greatly simplifies these calculations and enables a calculation of radiation damage as a standard deliverable in ITER analyses. (orig.)